How do you collaborate in a local area network?

I love Zotero for a lot of nice features and use it in my daily work.

BUT what me prevent to recommend and introduce this very nice tool in my company
is the lack to collaborate in local groups on a local zotero database (in an easily manner). That is an essential feature for me (and a lot of others).

I know, there exist since zotero 2.0b syncing and group features, and some hints
on the support pages and a lot of discussions on this topic (collaboration, syncing, groups) in the forums.

Until now I can't find a practicable solution for local collaborating in local groups
(multiple persons, multiple computers, multiple access at the same time, one common local database).

As I know, it is (until now) not possible to gather and share document meta informations, notes, tags and attachments with collegues inside a local network (without using the server). Nor is it on the roadmap of the zotero development team to provide this feature (please correct me, if this is wrong).

* My Question. How do people work around this problem at the moment (or in near future)? Tell me about your "solution".

(Do you waiting and hoping, that the feature will come?,
Do you use a version controlled zotero database?
Do you combine zotero with an addional reference management tool for collaborating (bibus, aigaion,..)?
Do you move from zotero to another tool?
Do you working on an implementation of a local zotero server?
Do you risk multiple access on the zotero database on a network drive?
  • Hi,

    it seems to me, that nobody in the zotero community has a practical solution for the above posted question or nobody collaborate on one common zotero database with collegues inside a LAN.

    I have switched to zotero a long time ago (it was a lot of work). I have done this
    because zotero it is a wonderfull single user reference managing tool AND I read on the homepage that a "Collaborate with group libraries" feature will come.
    Now this feature is available, but not in the kind as I hoped and as I need it.

    Perhaps somebody of the zotero-dev team can give a statement, whether or not zotero will go this direction ever. Else I have no chance to decide, whether or not zotero is the best tool for my intended use.

    Thanks for comments.
  • I have used Zotero in combination with refbase. Many options work if you are using BibTeX & can share the key system with coauthors. But if you're using Word and/or OO.o, I'd strongly recommend using the new groups functionality in order to maintain identifiers across multiple databases.

    I am not clear how your "version controlled zotero database" would work: merging is not trivial.

    It is not worth using Zotero on a network drive if you & your coworkers hope to be working at the same time.
  • Thanks for your reply.

    version controlled zotero was already a topic in the forums a long time ago (acutually I do not find the dicussion).

    My thougths about using subversion to control multiple read/write access are the following. Set the complete zotero folder including the sqlite database, the notes, the pdf's... in a version control repository. Every user can check out this folder to get a working copy of it and can work with in single user mode: read (including fulltext)/ write / delete / add contents to the database. When he finished work, he make a check in. And everybody else can now update the modifications.

    I think there exist not a lot of conflict situation because 2 users would (should) never modify the identical line of text. Mostly they will add content (additional notes, additional items, additional pdf's...) and will read what others wrote. This kind of usage should be a common agreement inside the group (That's practicable I think).

    What do you think about that way? Does this work in this szenario or do you see problems?

    The feature request of that above mentioned discussion was, to include a kind of subversion buttons to make it more easy to update, check in, check out a zotero database directly from the zotero gui.
  • But the sqlite database is a binary file. How is that going to be merged?
  • Hi noksagt,

    as I know, subversion is able to treat binary files with a special binary-diff algorithm. So it should be able to handle the sqlite database. See, where you can read this:

    "In the most general sense, Subversion handles binary files more gracefully than CVS does. Because CVS uses RCS, it can only store successive full copies of a changing binary file. But internally, Subversion expresses differences between files using a binary-differencing algorithm, regardless of whether they contain textual or binary data. That means that all files are stored differentially (compressed) in the repository, and small differences are always sent over the network.CVS users have to mark binary files with -kb flags, to prevent data from being garbled (due to keyword expansion and line-ending translations). They sometimes forget to do this."

    But I have never tried do use zotero with subversion, therefore I asked, whether or not somebody has tried it and has some experience how good and reliable it works.
  • That's just about reducing storage space by storing binary diffs. It doesn't help you merge database files.
  • I hoped, that subversion can handle the merging of binary files as long as no real conflict situation arise. So using subversion and preventing conflict situations by agreements inside the group, how to work with the zotero database, could be a workaround.

    But now it seems to me, that using subversion do not work as a workaround for local collaboration with zotero or at least you see some problems handling the binary sqlite database.

    Other solutions or workarounds for local collaboration?
  • I would really appreciate the possibility of collaborating on a local network rather than a webDAV server.

    I was unable to find a statement by the developers on a possible future implementation of this feature. Has any progress been done? Is that at all possible?

    Thanks for all the nice work,
  • Have anybody other solutions or workarounds for local collaboration?
  • One obvious solution would be to run your own server. That's not possible now because the server code has not been released.

    I believe the zotero team has indicated that the server code will be released at some point, but have not provided a timeline.
  • Hi zotero team,
    exist any timeline, when the zotero server code will be available to everybody?
  • Hi,

    for me it's a import question, which workarounds exist to collaborate in a local network. Until now, I got only a few comments on this topic:
    - Zotero and refbase works, but some zotero features disapear
    - version control do not work, because of the binary sqlite database
    - a local Zotero server is planned - timeline is not given

    Has nobody else in the zotero community a way to handle this problem?
  • Here's one possible way to integrate these use-cases:
    1. Release source code for a collaboration server

    2. Allow extension users to select multiple servers, with the default being

    3. Rework Groups so that each group is linked to a specific collaboration server

    I don't see any value in explicit LAN communication that could not be solved by running one's own (LAN-local) collab server.
  • Hi, we have the same situation. We would like to be able to place the database and document store on a network server with nothing on the local workstation except the plugins to FireFox and Word and maybe a configuration file.

    Is there a way to do this now? Will there be one in the near future?
  • it's really not possible atm.
    releasing the server code will solve some of the issues, and Dan has also indicated that file access on demand is planned at some points, but having no locally stored copies of the database might not be possible at all.
  • It seems like the current solution is to create a group on the server and have everyone create a custom storage location on the network server. Of course this means in a workgroup of 10 people we will have 10 copies of everything on the network server plus a copy of it all at When someone changes something the change will be recorded in their space on the network server, then synced to the server and then synced to the 9 other workstations and ultimately to the 9 other locations on our network server.

    Is this accurate?
  • yes, I think so.
    The alternative would be to work with links to documents (pointing to a location on your network) rather than attachments, along the lines suggested here:
  • Links work until you move or rename something. Still, this may be better -- and cheaper -- than the alternatives we have now. Thanks!
  • Another alternative (the one we're currently using) is to have a single account instead of a group. My workgroup of 5 has a single zotero account that gets synced to our respective local storage folders. Though I suppose you can't have a single network storage folder that can be simultaneously edited by multiple users--we each edit our local files and then sync to and thence to the other machines.

    Depending on how frequently users need access to the attached files, you could have them turn off file syncing, eliminating a copy of the stored files, and have access to the networked files storage via a smart folder/saved search.
Sign In or Register to comment.